Can I combine the 5% 1980 US Census with 1% 1980 US Cenus? The weights would be adjusted with perwt*5/6 etc.
Earlier posts said we can do this for 1970 Census. But 1980 Census is a 5% sample, would “sample overlap” become too big an issue that we are not supposed to merge a 5% with a 1% sample?
According to the “Selection of the Public-Use-Microdata Samples” section of the 1980 Sample Design page, the procedure for selecting the microdata samples was designed to, “minimize the likelihood that any one case would be selected into more than one public-use microdata sample, and the overlap among the samples may be considered negligible.” So combining the samples should not be problematic.
I hope this helps.